metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
- f1
model-index:
- name: flower_classification
results:
- task:
name: Image Classification
type: image-classification
dataset:
name: imagefolder
type: imagefolder
config: default
split: validation
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.9706601466992665
- name: F1
type: f1
value: 0.97382606978311
flower_classification
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.1638
- Accuracy: 0.9707
- F1: 0.9738
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0005
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_steps: 63
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 2.134 | 1.0 | 205 | 0.8454 | 0.8582 | 0.8377 |
| 0.6349 | 2.0 | 410 | 0.7229 | 0.8252 | 0.7947 |
| 0.3946 | 3.0 | 615 | 0.6453 | 0.8521 | 0.8301 |
| 0.2747 | 4.0 | 820 | 0.3665 | 0.9083 | 0.8901 |
| 0.1668 | 5.0 | 1025 | 0.3964 | 0.8998 | 0.8692 |
| 0.0767 | 6.0 | 1230 | 0.2997 | 0.9303 | 0.9282 |
| 0.0205 | 7.0 | 1435 | 0.1774 | 0.9584 | 0.9590 |
| 0.0066 | 8.0 | 1640 | 0.1467 | 0.9719 | 0.9732 |
| 0.0027 | 9.0 | 1845 | 0.1571 | 0.9707 | 0.9716 |
| 0.0026 | 10.0 | 2050 | 0.1603 | 0.9694 | 0.9709 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.2
- Datasets 2.18.0
- Tokenizers 0.15.2